{"id":"0220f976be0b15920a6fdeb0da99c3f773bfdfa6223b54e723fecd7aa65b8d9a","title":"FOX separates image-detection verdicts from explanations","argument":"Authors Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki and Symeon Papadopoulos introduce FOX, a system for detecting AI-generated images and producing forensic explanations. They argue that adapting a detector to a detection distribution can make it memorize generator-specific fingerprints, while jointly training classification and language generation can weaken the visual representation used for detection.\n\nFOX first builds and locks a DINOv3 detector, then uses its representation to guide a lightweight VLM in generating explanations. The authors’ central design constraint is that explanation training cannot alter the detector’s verdict. They report a mean AUC of 97.4 across 14 unseen generators, with BERTScore results of 69.8 on XPlainVerse and 56.1 on LOKI. These are reported benchmark results; the abstract does not establish performance in real-world use. How do the authors evaluate whether FOX’s explanations accurately reflect the evidence behind its verdicts?","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The supplied abstract reports benchmark results but does not describe the evaluation setup or establish performance in real-world use. The abstract does not explain how the accuracy of FOX's explanations is assessed.","speaker":{"name":"Giakoumoglou, Nikos, Giakoumoglou, Paschalis, Stathaki, Tania, Papadopoulos, Symeon","roleAtSource":"Article author"},"source":{"title":"Generalizable AI-Generated Image Detection with Embedding-Space Explanations","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.04111","publishedAt":"2026-11-03","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors introduce FOX to detect AI-generated images and provide forensic explanations while preventing explanation training from changing the detector's verdict.","excerpt":"We therefore require that explanation training cannot alter the verdict. We introduce FOX","url":"https://arxiv.org/abs/2611.04111","relationship":"Source-stated"},{"text":"The authors say detectors can generalize poorly when adapting a backbone to the detection distribution leads them to memorize generator-specific fingerprints.","excerpt":"generalize poorly because adapting a backbone on the detection distribution memorizes generator-specific fingerprints that quickly become obsolete.","url":"https://arxiv.org/abs/2611.04111","relationship":"Source-stated"},{"text":"FOX builds and locks a DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation.","excerpt":"FOX, which first builds and locks a specialist DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation.","url":"https://arxiv.org/abs/2611.04111","relationship":"Source-stated"},{"text":"The authors report mean AUC 97.4 across 14 unseen generators, and BERTScore results of 69.8 on XPlainVerse and 56.1 on LOKI.","excerpt":"FOX achieves 97.4 mean AUC across 14 unseen generators, 69.8 BERTScore on XPlainVerse, and 56.1 on LOKI.","url":"https://arxiv.org/abs/2611.04111","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.04111","checkedAt":"2026-11-10T06:26:07.143Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How do the authors evaluate whether FOX’s explanations accurately reflect the evidence behind its verdicts?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"FOX separates image-detection verdicts from explanations","summary":"Authors Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki and Symeon Papadopoulos introduce FOX, a system for detecting AI-generated images and producing forensic explanations. They argue that adapting a detector to a detection distribution can make it memorize generator-specific fingerprints, while jointly training classification and language generation can weaken the visual representation used for detection.\n\nFOX first builds and locks a DINOv3 detector, then uses its representation to guide a lightweight VLM in generating explanations. The authors’ central design constraint is that explanation training cannot alter the detector’s verdict. They report a mean AUC of 97.4 across 14 unseen generators, with BERTScore results of 69.8 on XPlainVerse and 56.1 on LOKI. These are reported benchmark results; the abstract does not establish performance in real-world use. How do the authors evaluate whether FOX’s explanations accurately reflect the evidence behind its verdicts?","question":"How do the authors evaluate whether FOX’s explanations accurately reflect the evidence behind its verdicts?","claims":[{"text":"The authors introduce FOX to detect AI-generated images and provide forensic explanations while preventing explanation training from changing the detector's verdict.","excerpt":"We therefore require that explanation training cannot alter the verdict. We introduce FOX","support":"source-stated"},{"text":"The authors say detectors can generalize poorly when adapting a backbone to the detection distribution leads them to memorize generator-specific fingerprints.","excerpt":"generalize poorly because adapting a backbone on the detection distribution memorizes generator-specific fingerprints that quickly become obsolete.","support":"source-stated"},{"text":"FOX builds and locks a DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation.","excerpt":"FOX, which first builds and locks a specialist DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation.","support":"source-stated"},{"text":"The authors report mean AUC 97.4 across 14 unseen generators, and BERTScore results of 69.8 on XPlainVerse and 56.1 on LOKI.","excerpt":"FOX achieves 97.4 mean AUC across 14 unseen generators, 69.8 BERTScore on XPlainVerse, and 56.1 on LOKI.","support":"source-stated"}],"limitations":["The supplied abstract reports benchmark results but does not describe the evaluation setup or establish performance in real-world use.","The abstract does not explain how the accuracy of FOX's explanations is assessed."],"mediaId":"d0d54bec046e89dfba40218722e581ba8a8e9753eb4ff5bc076143bf86d9b18a","correctionReason":null,"id":"0220f976be0b15920a6fdeb0da99c3f773bfdfa6223b54e723fecd7aa65b8d9a","publicationId":"666bcd2e-8444-4578-8578-27ec345d07ce","version":1,"sourceUrl":"https://arxiv.org/abs/2611.04111","sourceTitle":"Generalizable AI-Generated Image Detection with Embedding-Space Explanations","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-03","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-04T05:15:23.179Z","createdAt":"2026-11-10T05:23:19.309Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"d0d54bec046e89dfba40218722e581ba8a8e9753eb4ff5bc076143bf86d9b18a","kind":"article","url":"https://arxiv.org/abs/2611.04111","sourceUrl":"https://arxiv.org/abs/2611.04111","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-04T06:01:11.290Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.04111","checkedAt":"2026-11-10T06:26:07.143Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"d0d54bec046e89dfba40218722e581ba8a8e9753eb4ff5bc076143bf86d9b18a","confidence":0.86,"provider":"typesafe","model":"jev-1.13.0","inputHash":"1d1cf3a3e4817e0fc4b3a346029bfccf99963a191854734f069b91294b0d3f4b","decidedAt":"2026-11-04T06:01:12.250Z"},"provenance":{"textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observationId":"8390512c92ab96a70e5690313d5957ec190869a10d472f7525e20a4da6fe9e3f","retrievedAt":"2026-11-04T06:01:11.190Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"3b69d150-b31b-4d57-8297-a27750fd7275"},"reviewer":{"model":"gpt-6-luna","runId":"2383abaa-7ba6-4e79-8b81-e213ca81d0d7"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-10T05:23:19.309Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Giakoumoglou, Nikos","url":null,"kind":"unknown"},{"name":"Giakoumoglou, Paschalis","url":null,"kind":"unknown"},{"name":"Stathaki, Tania","url":null,"kind":"unknown"},{"name":"Papadopoulos, Symeon","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Giakoumoglou, Nikos","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Giakoumoglou, Nikos","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.04111","locator":"head/meta[name=author|citation_author]"},"assertionId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-10T06:26:07.143Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","sourceUrl":"https://arxiv.org/abs/2611.04111","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z"}},{"actor":{"name":"Giakoumoglou, Paschalis","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Giakoumoglou, Paschalis","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.04111","locator":"head/meta[name=author|citation_author]"},"assertionId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-10T06:26:07.143Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","sourceUrl":"https://arxiv.org/abs/2611.04111","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z"}},{"actor":{"name":"Stathaki, Tania","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Stathaki, Tania","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.04111","locator":"head/meta[name=author|citation_author]"},"assertionId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-10T06:26:07.143Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","sourceUrl":"https://arxiv.org/abs/2611.04111","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z"}},{"actor":{"name":"Papadopoulos, Symeon","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Papadopoulos, Symeon","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.04111","locator":"head/meta[name=author|citation_author]"},"assertionId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-10T06:26:07.143Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","sourceUrl":"https://arxiv.org/abs/2611.04111","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z"}}],"participationCoverage":{"total":4,"returned":4,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.04111","snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z","basis":"publisher-page-metadata","sourceLinks":44,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/0220f976be0b15920a6fdeb0da99c3f773bfdfa6223b54e723fecd7aa65b8d9a/source"}},"canonicalUrl":"https://escapemeridian.com/readings/fox-separates-image-detection-verdicts-from-explanations-0220f976","formats":{"html":"/readings/fox-separates-image-detection-verdicts-from-explanations-0220f976","markdown":"/readings/fox-separates-image-detection-verdicts-from-explanations-0220f976.md","json":"/readings/fox-separates-image-detection-verdicts-from-explanations-0220f976.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-benchmark","name":"Benchmark","explanation":"A defined set of tests used to compare systems. Results depend on the tasks and measures selected.","referenceUrl":"https://arxiv.org/abs/2211.09110"}],"entities":[{"id":"arxiv","name":"arXiv","kind":"organization","context":"arXiv is a research manuscript repository where authors share papers that may not yet have been peer reviewed.","referenceUrls":["https://arxiv.org/","https://info.arxiv.org/help/api/index.html"],"path":"/entities/arxiv","json":"/entities/arxiv.json","markdown":"/entities/arxiv.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"0220f976be0b15920a6fdeb0da99c3f773bfdfa6223b54e723fecd7aa65b8d9a","revision":"666bcd2e-8444-4578-8578-27ec345d07ce","revisionKind":"publication-id","title":"FOX separates image-detection verdicts from explanations","url":"https://escapemeridian.com/readings/fox-separates-image-detection-verdicts-from-explanations-0220f976"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.04111","revision":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","revisionKind":"sha256-source-text","title":"Generalizable AI-Generated Image Detection with Embedding-Space Explanations","url":"https://arxiv.org/abs/2611.04111"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."},"revisions":[{"id":"0220f976be0b15920a6fdeb0da99c3f773bfdfa6223b54e723fecd7aa65b8d9a","title":"FOX separates image-detection verdicts from explanations","argument":"Authors Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki and Symeon Papadopoulos introduce FOX, a system for detecting AI-generated images and producing forensic explanations. They argue that adapting a detector to a detection distribution can make it memorize generator-specific fingerprints, while jointly training classification and language generation can weaken the visual representation used for detection.\n\nFOX first builds and locks a DINOv3 detector, then uses its representation to guide a lightweight VLM in generating explanations. The authors’ central design constraint is that explanation training cannot alter the detector’s verdict. They report a mean AUC of 97.4 across 14 unseen generators, with BERTScore results of 69.8 on XPlainVerse and 56.1 on LOKI. These are reported benchmark results; the abstract does not establish performance in real-world use. How do the authors evaluate whether FOX’s explanations accurately reflect the evidence behind its verdicts?","topicIds":["intelligence-and-discovery"],"domain":"science","claimType":"attributed reading","limits":"The supplied abstract reports benchmark results but does not describe the evaluation setup or establish performance in real-world use. The abstract does not explain how the accuracy of FOX's explanations is assessed.","speaker":{"name":"Giakoumoglou, Nikos, Giakoumoglou, Paschalis, Stathaki, Tania, Papadopoulos, Symeon","roleAtSource":"Article author"},"source":{"title":"Generalizable AI-Generated Image Detection with Embedding-Space Explanations","publisher":"arXiv cs.AI","format":"Article","url":"https://arxiv.org/abs/2611.04111","publishedAt":"2026-11-03","dateBasis":"Date basis not provided."},"evidence":[{"text":"The authors introduce FOX to detect AI-generated images and provide forensic explanations while preventing explanation training from changing the detector's verdict.","excerpt":"We therefore require that explanation training cannot alter the verdict. We introduce FOX","url":"https://arxiv.org/abs/2611.04111","relationship":"Source-stated"},{"text":"The authors say detectors can generalize poorly when adapting a backbone to the detection distribution leads them to memorize generator-specific fingerprints.","excerpt":"generalize poorly because adapting a backbone on the detection distribution memorizes generator-specific fingerprints that quickly become obsolete.","url":"https://arxiv.org/abs/2611.04111","relationship":"Source-stated"},{"text":"FOX builds and locks a DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation.","excerpt":"FOX, which first builds and locks a specialist DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation.","url":"https://arxiv.org/abs/2611.04111","relationship":"Source-stated"},{"text":"The authors report mean AUC 97.4 across 14 unseen generators, and BERTScore results of 69.8 on XPlainVerse and 56.1 on LOKI.","excerpt":"FOX achieves 97.4 mean AUC across 14 unseen generators, 69.8 BERTScore on XPlainVerse, and 56.1 on LOKI.","url":"https://arxiv.org/abs/2611.04111","relationship":"Source-stated"}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.04111","checkedAt":"2026-11-10T06:26:07.143Z","items":[],"note":"No suitable media was found by the supported page extraction."},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"How do the authors evaluate whether FOX’s explanations accurately reflect the evidence behind its verdicts?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"FOX separates image-detection verdicts from explanations","summary":"Authors Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki and Symeon Papadopoulos introduce FOX, a system for detecting AI-generated images and producing forensic explanations. They argue that adapting a detector to a detection distribution can make it memorize generator-specific fingerprints, while jointly training classification and language generation can weaken the visual representation used for detection.\n\nFOX first builds and locks a DINOv3 detector, then uses its representation to guide a lightweight VLM in generating explanations. The authors’ central design constraint is that explanation training cannot alter the detector’s verdict. They report a mean AUC of 97.4 across 14 unseen generators, with BERTScore results of 69.8 on XPlainVerse and 56.1 on LOKI. These are reported benchmark results; the abstract does not establish performance in real-world use. How do the authors evaluate whether FOX’s explanations accurately reflect the evidence behind its verdicts?","question":"How do the authors evaluate whether FOX’s explanations accurately reflect the evidence behind its verdicts?","claims":[{"text":"The authors introduce FOX to detect AI-generated images and provide forensic explanations while preventing explanation training from changing the detector's verdict.","excerpt":"We therefore require that explanation training cannot alter the verdict. We introduce FOX","support":"source-stated"},{"text":"The authors say detectors can generalize poorly when adapting a backbone to the detection distribution leads them to memorize generator-specific fingerprints.","excerpt":"generalize poorly because adapting a backbone on the detection distribution memorizes generator-specific fingerprints that quickly become obsolete.","support":"source-stated"},{"text":"FOX builds and locks a DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation.","excerpt":"FOX, which first builds and locks a specialist DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation.","support":"source-stated"},{"text":"The authors report mean AUC 97.4 across 14 unseen generators, and BERTScore results of 69.8 on XPlainVerse and 56.1 on LOKI.","excerpt":"FOX achieves 97.4 mean AUC across 14 unseen generators, 69.8 BERTScore on XPlainVerse, and 56.1 on LOKI.","support":"source-stated"}],"limitations":["The supplied abstract reports benchmark results but does not describe the evaluation setup or establish performance in real-world use.","The abstract does not explain how the accuracy of FOX's explanations is assessed."],"mediaId":"d0d54bec046e89dfba40218722e581ba8a8e9753eb4ff5bc076143bf86d9b18a","correctionReason":null,"id":"0220f976be0b15920a6fdeb0da99c3f773bfdfa6223b54e723fecd7aa65b8d9a","publicationId":"666bcd2e-8444-4578-8578-27ec345d07ce","version":1,"sourceUrl":"https://arxiv.org/abs/2611.04111","sourceTitle":"Generalizable AI-Generated Image Detection with Embedding-Space Explanations","publisher":"arXiv cs.AI","publishedAt":null,"originalPublished":{"value":"2026-11-03","precision":"day","basis":"publisher_citation_date"},"firstObservedAt":"2026-11-04T05:15:23.179Z","createdAt":"2026-11-10T05:23:19.309Z","historical":true,"entityIds":["arxiv"],"topicIds":["intelligence-and-discovery"],"media":[{"id":"d0d54bec046e89dfba40218722e581ba8a8e9753eb4ff5bc076143bf86d9b18a","kind":"article","url":"https://arxiv.org/abs/2611.04111","sourceUrl":"https://arxiv.org/abs/2611.04111","label":"Read original","provenance":"source_link","verification":{"status":"verified","checkedAt":"2026-11-04T06:01:11.290Z","method":"acquired_source","contentType":"text/html"},"recommendationReason":"Recommended by Jev for understanding this source; checked by Luna."}],"sourceMedia":{"status":"not_found","sourceUrl":"https://arxiv.org/abs/2611.04111","checkedAt":"2026-11-10T06:26:07.143Z","items":[],"note":"No suitable media was found by the supported page extraction."},"editorialDecision":{"outcome":"accept","reasonCodes":["material_development"],"rubricVersion":"stream-admission-v5","mediaId":"d0d54bec046e89dfba40218722e581ba8a8e9753eb4ff5bc076143bf86d9b18a","confidence":0.86,"provider":"typesafe","model":"jev-1.13.0","inputHash":"1d1cf3a3e4817e0fc4b3a346029bfccf99963a191854734f069b91294b0d3f4b","decidedAt":"2026-11-04T06:01:12.250Z"},"provenance":{"textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observationId":"8390512c92ab96a70e5690313d5957ec190869a10d472f7525e20a4da6fe9e3f","retrievedAt":"2026-11-04T06:01:11.190Z","kind":"publisher-page","preparer":{"model":"gpt-6-luna","runId":"3b69d150-b31b-4d57-8297-a27750fd7275"},"reviewer":{"model":"gpt-6-luna","runId":"2383abaa-7ba6-4e79-8b81-e213ca81d0d7"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-10T05:23:19.309Z","model":"anthropic/claude-opus-5.5"},"sourceAttribution":{"authors":[{"name":"Giakoumoglou, Nikos","url":null,"kind":"unknown"},{"name":"Giakoumoglou, Paschalis","url":null,"kind":"unknown"},{"name":"Stathaki, Tania","url":null,"kind":"unknown"},{"name":"Papadopoulos, Symeon","url":null,"kind":"unknown"}],"authorRole":"article-author","hosts":[],"status":"stated","authorRoster":{"truncated":false,"omitted":0,"assertionGroupsOmitted":0},"participation":[{"actor":{"name":"Giakoumoglou, Nikos","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Giakoumoglou, Nikos","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.04111","locator":"head/meta[name=author|citation_author]"},"assertionId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f:participation:0","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-10T06:26:07.143Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","sourceUrl":"https://arxiv.org/abs/2611.04111","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z"}},{"actor":{"name":"Giakoumoglou, Paschalis","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Giakoumoglou, Paschalis","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.04111","locator":"head/meta[name=author|citation_author]"},"assertionId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f:participation:1","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-10T06:26:07.143Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","sourceUrl":"https://arxiv.org/abs/2611.04111","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z"}},{"actor":{"name":"Stathaki, Tania","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Stathaki, Tania","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.04111","locator":"head/meta[name=author|citation_author]"},"assertionId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f:participation:2","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-10T06:26:07.143Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","sourceUrl":"https://arxiv.org/abs/2611.04111","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z"}},{"actor":{"name":"Papadopoulos, Symeon","url":null,"kind":"unknown","actorId":null},"role":"author","scope":{"kind":"work-credit","locator":null},"literalCredit":"Papadopoulos, Symeon","status":"stated","provenance":{"basis":"head-meta","sourceUrl":"https://arxiv.org/abs/2611.04111","locator":"head/meta[name=author|citation_author]"},"assertionId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f:participation:3","identity":{"version":"entity-identity-v1","status":"unresolved","actorId":null,"canonicalActorId":null,"candidateActorIds":[],"assertions":[],"sourceObservedAt":"2026-11-10T06:26:07.143Z","knowledgeAsOf":null},"sourceVersion":{"snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","sourceUrl":"https://arxiv.org/abs/2611.04111","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z"}}],"participationCoverage":{"total":4,"returned":4,"complete":true,"fullAssertions":"source-metadata-endpoint"},"distribution":null,"publisher":null,"sourceUrl":"https://arxiv.org/abs/2611.04111","snapshotId":"f432e1d55e6825419019943797a8bbf447b6f4e35941748640c51654fa96356f","contentSha256":"a3923fd0eb29e6cdb3d84c5d37834c9be922f26ce2aacc6a5baef43b9413f8d1","textHash":"f61edab748105ce7c35199c64fd478a4d476acd7f0c57a27bef32d360d7265cd","observedAt":"2026-11-10T06:26:07.143Z","basis":"publisher-page-metadata","sourceLinks":44,"linksTruncated":false,"metadataUrl":"/api/pipeline/stream/0220f976be0b15920a6fdeb0da99c3f773bfdfa6223b54e723fecd7aa65b8d9a/source"}}}],"audits":[{"publicationId":"666bcd2e-8444-4578-8578-27ec345d07ce","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading correctly names the four listed authors and attributes the FOX design, motivation and benchmark figures to them as their own claims, not as established fact.","excerpt":"Authors:Nikos Giakoumoglou, Paschalis Giakoumoglou, Tania Stathaki, Symeon Papadopoulos"},{"dimension":"summary","status":"uncertain","explanation":"A non-specialist cannot follow the central terms. DINOv3, VLM, AUC, BERTScore, unseen generators and generator-specific fingerprints are never explained in plain language or flagged for tooltips. The reading also drops the abstract's stated stakes: realistic images, editors that leave no splice boundary, and black-box detectors with unverifiable verdicts. Without these, readers miss why explainable detection matters. Smallest repair: add one sentence on the black-box and editing problem, and gloss VLM, AUC and BERTScore (or add term-library entries).","excerpt":"Most detectors are black boxes with unverifiable verdicts and generalize poorly"},{"dimension":"date","status":"pass","explanation":"originalPublished 2026-11-03 matches the v1 submission date on the arXiv page.","excerpt":"[Submitted on 3 Nov 2026]"},{"dimension":"support","status":"pass","explanation":"The claims closely track the abstract. Saying joint training can weaken the visual representation is a fair paraphrase of the abstract's point that explainable methods forgo the generalization advantage because the representation is subjected to the language objective. The caveat that benchmarks do not establish real-world performance is appropriate. The question follows from the evidence without being leading.","excerpt":"explainable methods forgo this advantage: jointly training one multimodal model for classification and language generation subjects its visual representation to the language objective"},{"dimension":"media","status":"pass","explanation":"The selected media is the verified direct arXiv abstract page for this paper, which is the most direct access to the source.","excerpt":null}],"createdAt":"2026-11-10T05:23:19.309Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}